The proposed research will examine contexts of sex-typed communication between mothers and their preschool-aged (30-60 mos) children during play. Aspects of the immediate play setting as well as well the mothers' socio- cultural background will be considered. The first aim of the present application, then, is to research the influence of play activities on sex- typed mother-child communication. According to interactive models of gender (e.g., Deaux & Major, 1987), the immediate setting accounts for many observed behavioral gender differences. In this regard, play with sex- typed toys may be the source of some observed gender differences (Caldera et al., 1989). This will be tested in the first study. If the hypothesized play setting effects are confirmed, it would imply that children's skills and potential could be expanded through play with a greater diversity of toys and activities (cf. Serbin et al., 1977). A second aim of the proposed research is to consider how single-parenting is related to maternal sex-typed communication patterns. Research shows that children of single mothers are less sex-typed in their play preferences (Stevenson & Black, 1988). However, it is unknown if and how these differences might be reflected in mother-child interactions. The second study therefore will compare the communications of single and married mothers with their children. This study will also take into account the potentially confounding influence of maternal employment, which is associated with similar gender-related effects as single parenting (cf. Hoffman, 1989). Finally, a third aim of the research is to investigate the relation between ethnicity and SES-related factors on mothers' sex-typed play interactions with their children. Some reports suggest that patterns of sex-typing may vary between Mexican- and Anglo-American families (Staples & Mirande, 1980). Yet there have been no systematic observational studies of sex-typing with both Mexican- and Anglo-American mothers and their preschool-aged children. Therefore, the research will compare Chicano and Anglo mothers and children. Additionally, SES, acculturation, and extended family--which may mediate the effects of ethnicity (Vega, 1990)--will be taken into account. Using cultural variation as a "natural experiment," the research can help address theoretical questions about social influences on gender development. Also, the research will add to our knowledge about Mexican-American families.